Karpathy's autoresearch repo started an impressive trend. Agents can now train AI models to build ...
TL;DR · AI 摘要
Karpathy的autoresearch项目引发趋势，AI代理可训练模型构建最先进系统，但研究问题仍是关键。
核心要点
- AI代理能训练模型构建最先进系统
- 当前技术仅触及自动化研究的表面
- LLM在提出优质研究假设方面仍有局限
Agents can now train AI models to build SoTA agentic systems.
And to think this is just scratching the surface.
Ultimately, it boils down to good research questions or hypotheses. LLMs are not great at this (yet)." / X
elvis on X: "Karpathy's autoresearch repo started an impressive trend. Agents can now train AI models to build SoTA agentic systems. And to think this is just scratching the surface. Ultimately, it boils down to good research questions or hypotheses. LLMs are not great at this (yet)." / X
Don’t miss what’s happening
Karpathy's autoresearch repo started an impressive trend. Agents can now train AI models to build SoTA agentic systems. And to think this is just scratching the surface. Ultimately, it boils down to good research questions or hypotheses. LLMs are not great at this (yet).
Quote
Aksel
@akseljoonas
·
22h
Introducing ml-intern, the agent that just automated the post-training team @huggingface It's an open-source implementation of the real research loop that our ML researchers do every day. You give it a prompt, it researches papers, goes through citations, implements ideas in GPU
·
13
38
254
326
Read 13 replies